Many-query join: efficient shared execution of relational joins on modern hardware

  • Darko Makreshanski
  • Georgios Giannikis
  • Gustavo Alonso
  • Donald Kossmann
Special Issue Paper
  • 49 Downloads

Abstract

Database architectures typically process queries one at a time, executing concurrent queries in independent execution contexts. Often, such a design leads to unpredictable performance and poor scalability. One approach to circumvent the problem is to take advantage of sharing opportunities across concurrently running queries. In this paper, we propose many-query join (MQJoin), a novel method for sharing the execution of a join that can efficiently deal with hundreds of concurrent queries. This is achieved by minimizing redundant work and making efficient use of main-memory bandwidth and multi-core architectures. Compared to existing proposals, MQJoin is able to efficiently handle larger workloads regardless of the schema by exploiting more sharing opportunities. We also compared MQJoin to two commercial main-memory column-store databases. For a TPC-H-based workload, we show that MQJoin provides 2–5\(\times \) higher throughput with significantly more stable response times.

Keywords

RDBMS OLAP Analytics Join MQJoin Shared join Main memory TPC-H Xeon Phi MCDRAM 

References

  1. 1.
  2. 2.
    Albutiu, M.-C., Kemper, A., Neumann, T.: Massively parallel sort-merge joins in main memory multi-core database systems. PVLDB 5(10), 1064–1075 (2012)Google Scholar
  3. 3.
    Arumugam, S., Dobra, A., Jermaine, C.M., Pansare, N., Perez, L.: The DataPath system: a data-centric analytic processing engine for large data warehouses. Proc. SIGMOD 2010, 519–530 (2010)Google Scholar
  4. 4.
    Avnur, R., Hellerstein, J.M.: Eddies: continuously adaptive query processing. Proc. SIGMOD 2000, 261–272 (2000)Google Scholar
  5. 5.
    Balkesen, C., Alonso, G., Teubner, J., Özsu, M.T.: Multi-core, main-memory joins: sort versus hash revisited. PVLDB 7(1), 85–96 (2013)Google Scholar
  6. 6.
    Balkesen, C., Teubner, J., Alonso, G., Özsu, M.T.: Main-memory hash joins on multi-core CPUs: tuning to the underlying hardware. Proc. ICDE 2013, 362–373 (2013)Google Scholar
  7. 7.
    Balkesen, C., Teubner, J., Alonso, G., Özsu, T.: Main-memory hash joins on modern processor architectures. IEEE Trans. Knowl. Data Eng. 27(7), 1754–1766 (2015)CrossRefGoogle Scholar
  8. 8.
    Barber, R., Lohman, G., Pandis, I., Raman, V., Sidle, R., Attaluri, G., Chainani, N., Lightstone, S., Sharpe, D.: Memory-efficient hash joins. Proc. VLDB 8(4), 353–364 (2014)CrossRefGoogle Scholar
  9. 9.
    Blanas, S., Li, Y., Patel, J.M.: Design and evaluation of main memory hash join algorithms for multi-core CPUs. Proc. SIGMOD 2011, 37–48 (2011)Google Scholar
  10. 10.
    Boncz, P.A., Zukowski, M., Nes, N.: MonetDB/X100: hyper-pipelining query execution. Proc. CIDR 2005, 225–237 (2005)Google Scholar
  11. 11.
    Candea, G., Polyzotis, N., Vingralek, R.: A scalable, predictable join operator for highly concurrent data warehouses. PVLDB 2(1), 277–288 (2009)Google Scholar
  12. 12.
    Chen, C., Roussopoulos, N.: The implementation and performance evaluation of the ADMS query optimizer: integrating query result caching and matching. In: Proc EDBT, pp. 323–336 (1994)Google Scholar
  13. 13.
    Chen, S., Ailamaki, A., Gibbons, P. B., Mowry, T. C.: Improving hash join performance through prefetching. In: Proc. ICDE 2004, pp. 116– (2004)Google Scholar
  14. 14.
    Chen, S., Ailamaki, A., Gibbons, P.B., Mowry, T.C.: Improving hash join performance through prefetching. ACM Trans. Database Syst. 32(3), 17 (2007)CrossRefGoogle Scholar
  15. 15.
    Ebenstein, R., Kamat, N., Nandi, A.: FluxQuery: an execution framework for highly interactive query workloads. In: Proc SIGMOD, pp. 1333–1345. ACM, New York, NY, USA (2016)Google Scholar
  16. 16.
    Giannikis, G., Alonso, G., Kossmann, D.: SharedDB: killing one thousand queries with one stone. PVLDB 5(6), 526–537 (2012)Google Scholar
  17. 17.
    Giannikis, G., Makreshanski, D., Alonso, G., Kossmann, D.: Shared workload optimization. PVLDB 7(6), 429–440 (2014)Google Scholar
  18. 18.
    Graefe, G.: Volcano&#151 an extensible and parallel query evaluation system. IEEE Trans. Knowl. Data Eng. 6(1), 120–135 (1994)CrossRefGoogle Scholar
  19. 19.
    Harizopoulos, S., Ailamaki, A.: StagedDB: designing database servers for modern hardware. In: In IEEE Data, pp. 11–16 (2005)Google Scholar
  20. 20.
    Harizopoulos, S., Shkapenyuk, V., Ailamaki, A.: QPipe: a simultaneously pipelined relational query engine. Proc. SIGMOD 2005, 383–394 (2005)Google Scholar
  21. 21.
    Ivanova, M. G., Kersten, M. L., Nes, N. J., Gonçalves, R. A.: An architecture for recycling intermediates in a column-store. In: Proc. SIGMOD, pp. 309–320. ACM, New York, NY, USA (2009)Google Scholar
  22. 22.
    Jha, S., He, B., Lu, M., Cheng, X., Huynh, H.P.: Improving main memory hash joins on intel xeon phi processors: an experimental approach. PVLDB 8(6), 642–653 (2015)Google Scholar
  23. 23.
    Johnson, R., Harizopoulos, S., Hardavellas, N., Sabirli, K., Pandis, I., Ailamaki, A., Mancheril, N.G., Falsafi, B.: To share or not to share? Proc. VLDB 2007, 351–362 (2007)Google Scholar
  24. 24.
    Kim, C., Kaldewey, T., Lee, V.W., Sedlar, E., Nguyen, A.D., Satish, N., Chhugani, J., Di Blas, A., Dubey, P.: Sort versus hash revisited: fast join implementation on modern multi-core CPUs. PVLDB 2(2), 1378–1389 (2009)Google Scholar
  25. 25.
    Krikellas, K., Inc, G., Viglas, S. D., Cintra, M.: Modeling multithreaded query execution on chip multiprocessors. In ADMS (2010)Google Scholar
  26. 26.
    Lang, C.A., Bhattacharjee, B., Malkemus, T., Padmanabhan, S., Wong, K.: Increasing buffer-locality for multiple relational table scans through grouping and throttling. Proc. ICDE 2007, 1136–1145 (2007)Google Scholar
  27. 27.
    Lang, C. A., Bhattacharjee, B., Malkemus, T., Wong, K.: Increasing buffer-locality for multiple index based scans through intelligent placement and index scan speed control. In: Proc. VLDB, pp. 1298–1309 (2007)Google Scholar
  28. 28.
    Lang, H., Mühlbauer, T., Funke, F., Boncz, P. A., Neumann, T., Kemper, A.: Data blocks: hybrid OLTP and OLAP on compressed storage using both vectorization and compilation. In: Proceedings of the 2016 International Conference on Management of Data, SIGMOD ’16, pp. 311–326. ACM, New York, NY, USA (2016)Google Scholar
  29. 29.
    Larson, P.-A., Birka, A., Hanson, E.N., Huang, W., Nowakiewicz, M., Papadimos, V.: Real-time analytical processing with SQL server. Proc. VLDB 8(12), 1740–1751 (2015)CrossRefGoogle Scholar
  30. 30.
    Liu, F., Blanas, S.: Forecasting the cost of processing multi-join queries via hashing for main-memory databases. In Proc. SoCC, pp. 153–166. ACM, New York, NY, USA (2015)Google Scholar
  31. 31.
    Makreshanski, D., Giceva, J., Barthels, C., Alonso, G.: BatchDB: efficient isolated execution of hybrid OLTP+OLAP workloads for interactive applications. In: Proc. SIGMOD, pp. 37–50. ACM, New York, NY, USA (2017)Google Scholar
  32. 32.
    Manegold, S., Boncz, P., Kersten, M.: Optimizing main-memory join on modern hardware. IEEE Trans. Knowl. Data Eng. 14(4), 709–730 (2002)CrossRefGoogle Scholar
  33. 33.
    Manegold, S., Boncz, P., Kersten, M. L.: Generic database cost models for hierarchical memory systems. In: Proc VLDB, pp. 191–202. VLDB Endowment (2002)Google Scholar
  34. 34.
    Manegold, S., Pellenkoft, A., Kersten, M. L.: A multi-query optimizer for Monet. In: Proc. BNCOD, pp. 36–50. Springer, London, UK (2000)Google Scholar
  35. 35.
    Müller, I., Sanders, P., Lacurie, A., Lehner, W., Färber, F.: Cache-efficient aggregation: hashing is sorting. In: Proceedings of the 2015 ACM SIGMOD International Conference on Management of Data, Proc. SIGMOD 2015, pp. 1123–1136. ACM, New York, NY, USA (2015)Google Scholar
  36. 36.
    O’Neil, P., Graefe, G.: Multi-table joins through bitmapped join indices. SIGMOD Rec. 24(3), 8–11 (1995)CrossRefGoogle Scholar
  37. 37.
    O’Neil, P., O’Neal, B., Chen, X.: Star schema benchmark. http://www.cs.umb.edu/~poneil/StarSchemaB.PDF
  38. 38.
    Psaroudakis, I., Athanassoulis, M., Ailamaki, A.: Sharing data and work across concurrent analytical queries. PVLDB 6(9), 637–648 (2013)Google Scholar
  39. 39.
    Qiao, L., Raman, V., Reiss, F., Haas, P.J., Lohman, G.M.: Main-memory scan sharing for multi-core CPUs. PVLDB 1(1), 610–621 (2008)Google Scholar
  40. 40.
    Raman, V., Attaluri, G., Barber, R., Chainani, N., Kalmuk, D., KulandaiSamy, V., Leenstra, J., Lightstone, S., Liu, S., Lohman, G.M., Malkemus, T., Mueller, R., Pandis, I., Schiefer, B., Sharpe, D., Sidle, R., Storm, A., Zhang, L.: DB2 with BLU acceleration: so much more than just a column store. Proc. VLDB 6(11), 1080–1091 (2013)CrossRefGoogle Scholar
  41. 41.
    Raman, V., Swart, G., Qiao, L., Reiss, F., Dialani, V., Kossmann, D., Narang, I., Sidle, R.: Constant-time query processing. In: Proc. ICDE 2008, pp. 60–69 (2008)Google Scholar
  42. 42.
    Roy, P., Seshadri, S., Sudarshan, S., Bhobe, S.: Efficient and extensible algorithms for multi query optimization. In: Proc. SIGMOD, pp. 249–260. ACM, New York, NY, USA (2000)Google Scholar
  43. 43.
    Răducanu, B., Boncz, P., Zukowski, M.: Micro adaptivity in vectorwise. In: Proc. SIGMOD, pp. 1231–1242. ACM, New York, NY, USA (2013)Google Scholar
  44. 44.
    Sellis, T.K.: Multiple-query optimization. ACM Trans. Database Syst. 13(1), 23–52 (1988)CrossRefGoogle Scholar
  45. 45.
    Shatdal, A., Kant, C., Naughton, J.F.: Cache conscious algorithms for relational query processing. Proc. VLDB 1994, 510–521 (1994)Google Scholar
  46. 46.
    Sodani, A.: Knights landing (knl): 2nd generation intel(r) xeon phi processor. In: 2015 IEEE Hot Chips 27 Symposium (HCS), pp. 1–24 (Aug 2015)Google Scholar
  47. 47.
    Unterbrunner, P., Giannikis, G., Alonso, G., Fauser, D., Kossmann, D.: Predictable performance for unpredictable workloads. PVLDB 2(1), 706–717 (2009)Google Scholar
  48. 48.
    Valduriez, P.: Join indices. ACM Trans. Database Syst. 12(2), 218–246 (1987)CrossRefGoogle Scholar
  49. 49.
    Zukowski, M., Héman, S., Nes, N., Boncz, P.: Cooperative scans: dynamic bandwidth sharing in a DBMS. Proc. VLDB 2007, 723–734 (2007)Google Scholar
  50. 50.
    Zukowski, M., Nes, N., Boncz, P.: DSM versus NSM: CPU performance tradeoffs in block-oriented query processing. In: Proc. DaMoN 2008, pp. 47–54 (2008)Google Scholar
  51. 51.
    Zukowski, M., van de Wiel, M., Boncz, P.: Vectorwise: a vectorized analytical DBMS. Proc. ICDE 2012, 1349–1350 (2012)Google Scholar

Copyright information

© Springer-Verlag GmbH Germany 2017

Authors and Affiliations

  1. 1.Department of Computer ScienceETH ZurichZurichSwitzerland
  2. 2.Oracle Labs ZurichZurichSwitzerland
  3. 3.Microsoft ResearchRedmondUSA

Personalised recommendations